INDEX

Explanations

words related to urban environments or locations

oai_token-act-pair · gpt-3.5-turbo

references to urban environments or street settings

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

opsis

-0.77

estamp

-0.74

xit

-0.71

essee

-0.69

emort

-0.69

phabet

-0.64

auts

-0.63

itor

-0.62

inite

-0.62

izontal

-0.62

POSITIVE LOGITS

cape

1.09

 streets

1.06

ways

1.01

fare

0.93

 sidewalks

0.92

cars

0.91

walk

0.91

walker

0.89

 corners

0.84

worn

0.83

Activations Density 0.036%

words related to urban environments or locations

references to urban environments or street settings

No Comments

No Known Activations